Computer Vision for Understanding Catalyst Degradation Kinetics

Abstract

We report a computer vision strategy for the extraction and colorimetric analysis of catalyst degradation and product formation kinetics from video footage. The degradation of palladium(II) pre-catalyst systems to form ‘Pd black’ is investigated as a widely relevant case study for catalysis and materials chemistries. Beyond the study of catalysts in isolation, investigation of Pd-catalyzed Miyaura borylation reactions revealed informative correlations between colour parameters (most notably ΔE, a colour-agnostic measure of contrast change) and the concentration of product measured by off-line analysis (NMR and LC-MS). The breakdown of such correlations helped inform conditions under which reaction vessels were compromised by air ingress. These findings present opportunities to expand the toolbox of non-invasive analytical techniques, operationally cheaper and simpler to implement than common spectroscopic methods. The approach introduces the capability of analyzing the macroscopic ‘bulk’ for the study of reaction kinetics in complex mixtures, in complement to the more common study of microscopic and molecular specifics.

Computer Vision for Kinetic Analysis of Lab- and Process-Scale Mixing Phenomena

“…In practice, Δ Ε serves as a color-agnostic measure of contrast over time, as measured relative to the color recorded at time zero. In earlier work, we applied ΔΕ -time profiles to effectively capture productive and degradative processes in palladium-catalyzed reactions . The same metric has found limited applications in powder mixing analysis using a progressive single image analysis approach. , Here, we enable full video-based Δ Ε analysis, both as a grand average of all color progress across the entire bulk of a vessel and as a spatially resolved cell-based analysis to capture meso-mixing phenomena in an intuitive manner.…”

“…The tool has been developed for end users in small-scale chemical development and intermediate-scale process chemistry and collaboratively designed with industrial chemists working in these domains. Following earlier applications of this technology in small-scale electrochemical reaction development 45 and palladium catalyst degradation kinetics, 46 the present study focused on extending the application of Kineticolor to mixing analysis. In other words, where previous studies focused primarily on overall reaction bulk and color averaging, the present study more deeply explored spatially resolved image analysis methods.…”

Computer Vision for Kinetic Analysis of Lab- and Process-scale Mixing Phenomena

“…Following earlier applications of this technology in small scale electrochemical reaction development 43 and palladium catalyst degradation kinetics, 44 the present study focused on extending the application of Kineticolor to mixing analysis. In other words, where previous studies 5 focused primarily on overall reaction bulk and color averaging, the present study more deeply explored spatially-resolved image analysis methods.…”

“…We herein more deeply profiles to effectively capture productive and degradative processes in palladium-catalyzed reactions. 44 The same metric has found limited applications in powder mixing analysis using a progressive single image analysis approach. 50,51 Here, we enable full video-based E analysis, both as a grand average of all color progress across the entire bulk of a vessel, and as a spatially Homogeneity somewhat tracked ASM, being designed to measure whole image similarity.…”

“…We concluded the study by investigating the predictive power to determine relative product/reactant ratios using the colorimetric data extracted with Kineticolor. As per our earlier study investigating palladium-catalyzed borylation processes, 44 we applied Shannon's Mutual Information (M.I.) 58…”